---
license: cc-by-nc-4.0
language:
- zh
- en
base_model:
- meta-llama/Llama-3.2-3B-Instruct
tags:
- Text-to-Speech
pipeline_tag: text-to-speech
---

[![arXiv](https://img.shields.io/badge/arXiv-Paper-<COLOR>.svg)](https://arxiv.org/abs/2502.04128)

**Update (2025-02-13):** Add [Llasa finetune instruction](https://github.com/zhenye234/LLaSA_training/tree/main/finetune).

These models are not mentioned in the original paper, they are essentially the same as LLaSA 1B and LLaSA 3B, except they have been fine-tuned with a mixed speech and text SFT dataset, which enables the model to retain text-based conversational abilities.
 
LLaSA: Scaling Train-Time and Inference-Time Compute for LLaMA-based Speech Synthesis